News for hackers. Stuff that matters.
Re:AI competitions don’t produce useful models by emmab on Friday September 20, 2019 @01:00AM (#21023449)
https://www.lesswrong.com/posts/5gQLrJr2yhPzMCcni/the-optimi...

The competition and test set will still have hidden biases due to the ontology you use.

Sufficient optimization pressure always eventually overcomes your bias control metrics, in the context of actual utility rather than other metrics.

It's well known in cryptography and security that all abstractions are leaky.

There are no comments on this comment yet.